What are the current views on last and corresponding authorship in ecology?

Introduction

Who is the last author on a paper? Is it the person who did the least work? Or is it the PI of the lab where the work was done? When I started grad school (in 2000), the norm in ecology was still that the last author on a paper is the person who did the least work. But, more recently, it has seemed to me that the norm is that the last author on a paper is the “senior” author (usually the PI). However, if you talk with other ecologists about the topic, it’s clear that there’s variation in views, and that not everyone is on the same page.

This project started out as a poll on the Dynamic Ecology blog, but has led to a bigger project. The code here has been updated to reflect what is in the manuscript on this (in revision for Ecology & Evolution), rather than the original blog posts.

Literature review

I used a combination of Web of Science data and manually searching through journals to determine the number of authors and corresponding authorship for papers in Ecology from 1956-2016 (every 10 years from 1956-1996, every five years from 2001-2016) and in American Naturalist, Evolution, and Oikos in 2001, 2006, 2011, and 2016.

Making figures and doing analyses for revised version of manuscript (incorporating suggestions from the reviewers)

## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
## 
## Attaching package: 'cowplot'
## The following object is masked from 'package:ggplot2':
## 
##     ggsave

## 
## Call:
## glm(formula = numberauthors ~ Year, family = poisson(), data = ecologyWoSdata)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.9081  -0.6508  -0.1922   0.3472   8.5998  
## 
## Coefficients:
##              Estimate Std. Error z value Pr(>|z|)    
## (Intercept) -47.87826    2.00366  -23.89   <2e-16 ***
## Year          0.02447    0.00100   24.46   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for poisson family taken to be 1)
## 
##     Null deviance: 2531.1  on 1911  degrees of freedom
## Residual deviance: 1822.8  on 1910  degrees of freedom
## AIC: 7146
## 
## Number of Fisher Scoring iterations: 5
## 
## Call:
## glm(formula = numberauthors ~ Year + Journal + Year * Journal, 
##     family = poisson(), data = recentWoSdata)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.0500  -0.7499  -0.3020   0.3934   8.6805  
## 
## Coefficients:
##                         Estimate Std. Error z value Pr(>|z|)    
## (Intercept)           -46.747359   8.743948  -5.346 8.98e-08 ***
## Year                    0.023834   0.004351   5.477 4.31e-08 ***
## JournalEcology        -22.925916  10.334102  -2.218   0.0265 *  
## JournalEvolution      -12.202433  10.870768  -1.122   0.2617    
## JournalOikos          -23.118439  11.177112  -2.068   0.0386 *  
## Year:JournalEcology     0.011487   0.005142   2.234   0.0255 *  
## Year:JournalEvolution   0.006102   0.005410   1.128   0.2593    
## Year:JournalOikos       0.011539   0.005563   2.074   0.0380 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for poisson family taken to be 1)
## 
##     Null deviance: 4078.2  on 3619  degrees of freedom
## Residual deviance: 3647.5  on 3612  degrees of freedom
## AIC: 14369
## 
## Number of Fisher Scoring iterations: 5
## Analysis of Deviance Table
## 
## Model 1: numberauthors ~ Year + Journal + Year * Journal
## Model 2: numberauthors ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance Pr(>Chi)  
## 1      3612     3647.5                       
## 2      3615     3653.8 -3  -6.3166  0.09718 .
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
## 
## Model 1: numberauthors ~ Year
## Model 2: numberauthors ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
## 1      3618     3692.8                          
## 2      3615     3653.8  3   39.014 1.724e-08 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
## 
## Model 1: numberauthors ~ Journal
## Model 2: numberauthors ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
## 1      3616     4038.2                          
## 2      3615     3653.8  1   384.34 < 2.2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## # A tibble: 9 x 3
##    Year `median(numberauthors)` `mean(numberauthors)`
##   <int>                   <dbl>                 <dbl>
## 1  1956                       1              1.378049
## 2  1966                       1              1.517241
## 3  1976                       1              1.663934
## 4  1986                       2              1.756250
## 5  1996                       2              2.292887
## 6  2001                       2              2.688406
## 7  2006                       3              3.255255
## 8  2011                       3              4.030435
## 9  2016                       4              4.566154
## # A tibble: 5 x 3
##   Correspondence     n rel.freq
##           <fctr> <int>    <dbl>
## 1            all     2        0
## 2          first   751       84
## 3           last   114       13
## 4         middle    16        2
## 5          other     8        1
## Warning in bind_rows_(x, .id): Unequal factor levels: coercing to character
## Warning in bind_rows_(x, .id): binding character and factor vector,
## coercing into character vector

## Warning in bind_rows_(x, .id): binding character and factor vector,
## coercing into character vector

## 
## Call:
## glm(formula = Cor01 ~ Year + Journal + Year * Journal, family = binomial(), 
##     data = recentWoSdataCor)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.0991   0.4838   0.5142   0.5975   1.4107  
## 
## Coefficients:
##                         Estimate Std. Error z value Pr(>|z|)    
## (Intercept)           -1.910e+02  4.327e+01  -4.414 1.02e-05 ***
## Year                   9.568e-02  2.154e-02   4.442 8.93e-06 ***
## JournalEcology         1.494e+02  5.773e+01   2.587  0.00967 ** 
## JournalEvolution      -1.156e+01  5.308e+01  -0.218  0.82764    
## JournalOikos           1.823e+02  6.395e+01   2.851  0.00436 ** 
## Year:JournalEcology   -7.406e-02  2.873e-02  -2.578  0.00994 ** 
## Year:JournalEvolution  5.281e-03  2.642e-02   0.200  0.84158    
## Year:JournalOikos     -9.034e-02  3.183e-02  -2.838  0.00454 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 3159.0  on 2984  degrees of freedom
## Residual deviance: 2841.7  on 2977  degrees of freedom
## AIC: 2857.7
## 
## Number of Fisher Scoring iterations: 4
## Analysis of Deviance Table
## 
## Model 1: Cor01 ~ Year + Journal + Year * Journal
## Model 2: Cor01 ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
## 1      2977     2841.7                          
## 2      2980     2861.1 -3  -19.344 0.0002321 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
## 
## Model 1: Cor01 ~ Year
## Model 2: Cor01 ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
## 1      2983     3119.9                          
## 2      2980     2861.1  3   258.88 < 2.2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
## 
## Model 1: Cor01 ~ Journal
## Model 2: Cor01 ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
## 1      2981     2909.1                          
## 2      2980     2861.1  1   48.013 4.235e-12 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Call:
## glm(formula = LastCor01 ~ Year + Journal + Year * Journal, family = binomial(), 
##     data = recentWoSdataCor)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -0.6107  -0.4933  -0.4215  -0.3799   2.4560  
## 
## Coefficients:
##                         Estimate Std. Error z value Pr(>|z|)  
## (Intercept)           -1.347e+02  6.808e+01  -1.978   0.0479 *
## Year                   6.585e-02  3.384e-02   1.946   0.0517 .
## JournalEcology         2.592e+00  8.324e+01   0.031   0.9752  
## JournalEvolution      -5.252e+01  8.458e+01  -0.621   0.5346  
## JournalOikos           1.065e+02  8.711e+01   1.222   0.2217  
## Year:JournalEcology   -1.291e-03  4.138e-02  -0.031   0.9751  
## Year:JournalEvolution  2.623e-02  4.205e-02   0.624   0.5328  
## Year:JournalOikos     -5.299e-02  4.332e-02  -1.223   0.2213  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1889.7  on 2984  degrees of freedom
## Residual deviance: 1859.0  on 2977  degrees of freedom
## AIC: 1875
## 
## Number of Fisher Scoring iterations: 5
## Analysis of Deviance Table
## 
## Model 1: LastCor01 ~ Year + Journal + Year * Journal
## Model 2: LastCor01 ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1      2977     1859.0                     
## 2      2980     1863.8 -3  -4.7582   0.1904
## Analysis of Deviance Table
## 
## Model 1: LastCor01 ~ Year
## Model 2: LastCor01 ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1      2983     1867.4                     
## 2      2980     1863.8  3   3.6155   0.3061
## Analysis of Deviance Table
## 
## Model 1: LastCor01 ~ Journal
## Model 2: LastCor01 ~ Year + Journal
##   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
## 1      2981     1885.7                          
## 2      2980     1863.8  1   21.929 2.829e-06 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## # A tibble: 11 x 4
## # Groups:   Region [6]
##           Region LastCor01     n rel.freq
##           <fctr>     <dbl> <int>    <dbl>
##  1        Africa         0    10      100
##  2          Asia         0    33       63
##  3          Asia         1    19       37
##  4        Europe         0   255       86
##  5        Europe         1    41       14
##  6 North America         0   398       91
##  7 North America         1    40        9
##  8       Oceania         0    60       86
##  9       Oceania         1    10       14
## 10 South America         0    21       84
## 11 South America         1     4       16
## # A tibble: 11 x 4
## # Groups:   Region [6]
##           Region Cor01     n rel.freq
##           <fctr> <dbl> <int>    <dbl>
##  1        Africa     1    10      100
##  2          Asia     0    22       42
##  3          Asia     1    30       58
##  4        Europe     0    49       17
##  5        Europe     1   247       83
##  6 North America     0    53       12
##  7 North America     1   385       88
##  8       Oceania     0    11       16
##  9       Oceania     1    59       84
## 10 South America     0     6       24
## 11 South America     1    19       76

## 
## Call:
## glm(formula = LastCor01 ~ Region, family = binomial(link = logit), 
##     data = WoSdata2016EurNA)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -0.5461  -0.5461  -0.4376  -0.4376   2.1879  
## 
## Coefficients:
##                     Estimate Std. Error z value Pr(>|z|)    
## (Intercept)          -1.8277     0.1683 -10.862   <2e-16 ***
## RegionNorth America  -0.4699     0.2363  -1.989   0.0467 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 509.77  on 733  degrees of freedom
## Residual deviance: 505.83  on 732  degrees of freedom
## AIC: 509.83
## 
## Number of Fisher Scoring iterations: 5
## 
## Call:
## glm(formula = Cor01 ~ Region, family = binomial(link = logit), 
##     data = WoSdata2016EurNA)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.0552   0.5079   0.5079   0.6016   0.6016  
## 
## Coefficients:
##                     Estimate Std. Error z value Pr(>|z|)    
## (Intercept)           1.6176     0.1564  10.343   <2e-16 ***
## RegionNorth America   0.3654     0.2143   1.705   0.0882 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 591.72  on 733  degrees of freedom
## Residual deviance: 588.83  on 732  degrees of freedom
## AIC: 592.83
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastCor01 ~ Region, family = binomial(link = logit), 
##     data = WoSdata2016RegionsEnough)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -0.9537  -0.5461  -0.4376  -0.4376   2.1879  
## 
## Coefficients:
##                     Estimate Std. Error z value Pr(>|z|)    
## (Intercept)          -0.5521     0.2880  -1.917 0.055236 .  
## RegionEurope         -1.2756     0.3335  -3.825 0.000131 ***
## RegionNorth America  -1.7455     0.3323  -5.252  1.5e-07 ***
## RegionOceania        -1.2397     0.4468  -2.775 0.005524 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 656.61  on 855  degrees of freedom
## Residual deviance: 631.52  on 852  degrees of freedom
## AIC: 639.52
## 
## Number of Fisher Scoring iterations: 5
## 
## Call:
## glm(formula = Cor01 ~ Region, family = binomial(link = logit), 
##     data = WoSdata2016RegionsEnough)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.0552   0.5079   0.5079   0.6016   1.0489  
## 
## Coefficients:
##                     Estimate Std. Error z value Pr(>|z|)    
## (Intercept)           0.3102     0.2807   1.105  0.26917    
## RegionEurope          1.3074     0.3213   4.069 4.72e-05 ***
## RegionNorth America   1.6728     0.3166   5.283 1.27e-07 ***
## RegionOceania         1.3695     0.4320   3.170  0.00152 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 746.18  on 855  degrees of freedom
## Residual deviance: 720.57  on 852  degrees of freedom
## AIC: 728.57
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = Cor01 ~ ordered(binnedauthors), family = binomial(link = "logit"), 
##     data = WoSdata2016RegionsNot1)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.2101   0.4880   0.5729   0.6681   0.8257  
## 
## Coefficients:
##                           Estimate Std. Error z value Pr(>|z|)
## (Intercept)                3.12201   48.75779   0.064    0.949
## ordered(binnedauthors).L   7.18002  226.60539   0.032    0.975
## ordered(binnedauthors).Q   7.80157  233.37106   0.033    0.973
## ordered(binnedauthors).C   6.67405  195.25234   0.034    0.973
## ordered(binnedauthors)^4   3.68644  137.30401   0.027    0.979
## ordered(binnedauthors)^5   2.19802   81.13878   0.027    0.978
## ordered(binnedauthors)^6   0.57633   39.44861   0.015    0.988
## ordered(binnedauthors)^7   0.17907   14.98395   0.012    0.990
## ordered(binnedauthors)^8  -0.01144    3.87615  -0.003    0.998
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 761.25  on 842  degrees of freedom
## Residual deviance: 743.49  on 834  degrees of freedom
## AIC: 761.49
## 
## Number of Fisher Scoring iterations: 14
## 
## Call:
## glm(formula = LastCor01 ~ ordered(binnedauthors), family = binomial(link = "logit"), 
##     data = WoSdata2016RegionsNot1)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -0.6681  -0.5815  -0.5389  -0.4516   2.2101  
## 
## Coefficients:
##                            Estimate Std. Error z value Pr(>|z|)
## (Intercept)               -3.346513  48.757801  -0.069    0.945
## ordered(binnedauthors).L  -7.054922 226.605421  -0.031    0.975
## ordered(binnedauthors).Q  -7.573393 233.371092  -0.032    0.974
## ordered(binnedauthors).C  -6.418017 195.252382  -0.033    0.974
## ordered(binnedauthors)^4  -3.801151 137.304047  -0.028    0.978
## ordered(binnedauthors)^5  -2.305563  81.138836  -0.028    0.977
## ordered(binnedauthors)^6  -0.726725  39.448764  -0.018    0.985
## ordered(binnedauthors)^7  -0.353257  14.984429  -0.024    0.981
## ordered(binnedauthors)^8  -0.002962   3.877852  -0.001    0.999
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 668.01  on 842  degrees of freedom
## Residual deviance: 658.55  on 834  degrees of freedom
## AIC: 676.55
## 
## Number of Fisher Scoring iterations: 14
## 
## Call:
## glm(formula = Cor01 ~ ordered(binnedauthors7), family = binomial(link = "logit"), 
##     data = WoSdata2016RegionsNot1)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.0915   0.4880   0.5729   0.6559   0.7225  
## 
## Coefficients:
##                           Estimate Std. Error z value Pr(>|z|)    
## (Intercept)                1.55345    0.09442  16.452   <2e-16 ***
## ordered(binnedauthors7).L -0.58424    0.23106  -2.528   0.0115 *  
## ordered(binnedauthors7).Q  0.32459    0.22584   1.437   0.1506    
## ordered(binnedauthors7).C  0.07147    0.23535   0.304   0.7614    
## ordered(binnedauthors7)^4  0.09013    0.23660   0.381   0.7033    
## ordered(binnedauthors7)^5  0.03278    0.22737   0.144   0.8854    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 761.25  on 842  degrees of freedom
## Residual deviance: 752.61  on 837  degrees of freedom
## AIC: 764.61
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastCor01 ~ ordered(binnedauthors7), family = binomial(link = "logit"), 
##     data = WoSdata2016RegionsNot1)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -0.6216  -0.5601  -0.5389  -0.4516   2.1604  
## 
## Coefficients:
##                           Estimate Std. Error z value Pr(>|z|)    
## (Intercept)               -1.81604    0.10330 -17.580   <2e-16 ***
## ordered(binnedauthors7).L  0.40227    0.25239   1.594    0.111    
## ordered(binnedauthors7).Q -0.29073    0.24791  -1.173    0.241    
## ordered(binnedauthors7).C -0.03314    0.25671  -0.129    0.897    
## ordered(binnedauthors7)^4 -0.14677    0.25806  -0.569    0.570    
## ordered(binnedauthors7)^5  0.02015    0.24997   0.081    0.936    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 668.01  on 842  degrees of freedom
## Residual deviance: 663.76  on 837  degrees of freedom
## AIC: 675.76
## 
## Number of Fisher Scoring iterations: 4

## # A tibble: 10 x 2
##    binnedauthors     n
##           <fctr> <int>
##  1             1    48
##  2           10+    45
##  3             2   196
##  4             3   185
##  5             4   150
##  6             5   114
##  7             6    74
##  8             7    45
##  9             8    23
## 10             9    11
## # A tibble: 10 x 2
##    binnedauthors     n
##           <fctr> <int>
##  1             1    48
##  2           10+    45
##  3             2   196
##  4             3   185
##  5             4   150
##  6             5   114
##  7             6    74
##  8             7    45
##  9             8    23
## 10             9    11
## # A tibble: 18 x 4
## # Groups:   binnedauthors [10]
##    binnedauthors LastCor01     n rel.freq
##           <fctr>     <dbl> <int>    <dbl>
##  1             1         0    48      100
##  2           10+         0    38       84
##  3           10+         1     7       16
##  4             2         0   177       90
##  5             2         1    19       10
##  6             3         0   160       86
##  7             3         1    25       14
##  8             4         0   129       86
##  9             4         1    21       14
## 10             5         0    96       84
## 11             5         1    18       16
## 12             6         0    61       82
## 13             6         1    13       18
## 14             7         0    36       80
## 15             7         1     9       20
## 16             8         0    21       91
## 17             8         1     2        9
## 18             9         0    11      100
## # A tibble: 18 x 4
## # Groups:   binnedauthors [10]
##    binnedauthors Cor01     n rel.freq
##           <fctr> <dbl> <int>    <dbl>
##  1             1     1    48      100
##  2           10+     0     9       20
##  3           10+     1    36       80
##  4             2     0    22       11
##  5             2     1   174       89
##  6             3     0    28       15
##  7             3     1   157       85
##  8             4     0    27       18
##  9             4     1   123       82
## 10             5     0    23       20
## 11             5     1    91       80
## 12             6     0    17       23
## 13             6     1    57       77
## 14             7     0    13       29
## 15             7     1    32       71
## 16             8     0     2        9
## 17             8     1    21       91
## 18             9     1    11      100
## # A tibble: 13 x 4
## # Groups:   binnedauthors7 [7]
##    binnedauthors7 LastCor01     n rel.freq
##            <fctr>     <dbl> <int>    <dbl>
##  1              1         0    48      100
##  2              2         0   177       90
##  3              2         1    19       10
##  4              3         0   160       86
##  5              3         1    25       14
##  6              4         0   129       86
##  7              4         1    21       14
##  8              5         0    96       84
##  9              5         1    18       16
## 10              6         0    61       82
## 11              6         1    13       18
## 12             7+         0   106       85
## 13             7+         1    18       15
## # A tibble: 13 x 4
## # Groups:   binnedauthors7 [7]
##    binnedauthors7 Cor01     n rel.freq
##            <fctr> <dbl> <int>    <dbl>
##  1              1     1    48      100
##  2              2     0    22       11
##  3              2     1   174       89
##  4              3     0    28       15
##  5              3     1   157       85
##  6              4     0    27       18
##  7              4     1   123       82
##  8              5     0    23       20
##  9              5     1    91       80
## 10              6     0    17       23
## 11              6     1    57       77
## 12             7+     0    24       19
## 13             7+     1   100       81

The poll

The poll had four main questions:

  1. For ecology papers, do you consider the last author to be the senior author?
  2. Which of the following statements most closely matches the current norms in ecology in terms of who is corresponding author?
  3. Which of the following statements would be best practice in terms of who is corresponding author?
  4. If someone includes a statement on his/her CV indicating they have used a first/last author emphasis, do you pay attention to that?

It also asked about the respondent’s primary research area, whether their research is primarily basic or applied, how frequently they conduct interdisciplinary research, how many years post-PhD they are, where they live, and what their current department is.

The poll first appeared on 6 April 2016 and ran for two weeks.

Basic overview of responses

After removing the four blank responses, there were 1122 responses to the poll. What did the respondents look like?

Primary Research Area of Respondents
PrimaryResearch n rel.freq
Biology other than EEB 24 2
Ecology (primarily computational-based) 217 19
Ecology (primarily field-based) 558 50
Ecology (primarily wet-lab based, including molecular ecology) 119 11
Evolutionary biology (primarily molecular) 51 5
Evolutionary biology (primarily organismal) 130 12
Outside biology 21 2
BasicApplied n rel.freq
Applied 362 33
Basic 751 67
Interdisciplinary n rel.freq
Always 50 4
Often 271 24
Sometimes 401 36
Rarely 293 26
Never 99 9
YearssincePhD n rel.freq
0 (current students should choose this) 311 28
5-Jan 344 31
10-Jun 200 18
15-Nov 136 12
16-20 57 5
>20 53 5
I do not have a PhD and am not a current student 20 2
WhereLive n rel.freq
Africa 8 1
Asia 13 1
Australia 63 6
Europe 288 26
North America 717 64
South America 30 3
Dept01 n rel.freq
1 304 28
2 444 41
3 212 19
4 134 12

Results for the four main questions

Q1: “For ecology papers, do you consider the last author to be the senior author?”

## # A tibble: 6 x 3
##   LastSenior01     n rel.freq
##          <int> <int>    <dbl>
## 1            1    77        7
## 2            2    61        5
## 3            3    19        2
## 4            4    88        8
## 5            5   395       35
## 6            6   480       43
## [1] 1 2 3 4 5 6
## Levels: 1 2 3 4 5 6

Q2: “Which of the following statements most closely matches the current norms in ecology in terms of who is corresponding author?”

## # A tibble: 5 x 3
##   CorrespondingCurrent01     n rel.freq
##                    <int> <int>    <dbl>
## 1                      1   182       16
## 2                      2    37        3
## 3                      3   215       19
## 4                      4    82        7
## 5                      5   602       54

Q3: “Which of the following statements would be best practice in terms of who is corresponding author?”

## # A tibble: 5 x 3
##   CorrespondingBest01     n rel.freq
##                 <int> <int>    <dbl>
## 1                   1   266       24
## 2                   2    40        4
## 3                   3    84        8
## 4                   4    46        4
## 5                   5   676       61

Q4: “If someone includes a statement on his/her CV indicating they have used a first/last author emphasis, do you pay attention to that?”

## # A tibble: 4 x 3
##   CVStatement01     n rel.freq
##           <int> <int>    <dbl>
## 1             1    87        8
## 2             2   234       21
## 3             3   538       49
## 4             4   245       22

Looking at cross-tabs

Does whether people view the last author as the senior author vary based on age, country, research area, and/or department?

## Scale for 'y' is already present. Adding another scale for 'y', which
## will replace the existing scale.

## Scale for 'y' is already present. Adding another scale for 'y', which
## will replace the existing scale.
## Scale for 'y' is already present. Adding another scale for 'y', which
## will replace the existing scale.
## Scale for 'y' is already present. Adding another scale for 'y', which
## will replace the existing scale.

## 
## Call:
## glm(formula = LastSeniorYes ~ ordered(PhD01), family = binomial(link = "logit"), 
##     data = polldatacareerstageanalysis)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.8519   0.6300   0.6527   0.6608   0.8758  
## 
## Coefficients:
##                  Estimate Std. Error z value Pr(>|z|)    
## (Intercept)       1.19334    0.09081  13.142   <2e-16 ***
## ordered(PhD01).L -0.51899    0.23563  -2.203   0.0276 *  
## ordered(PhD01).Q -0.05427    0.21699  -0.250   0.8025    
## ordered(PhD01).C  0.13817    0.23170   0.596   0.5510    
## ordered(PhD01)^4 -0.20785    0.22774  -0.913   0.3614    
## ordered(PhD01)^5 -0.43600    0.19804  -2.202   0.0277 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1140.7  on 1098  degrees of freedom
## Residual deviance: 1127.4  on 1093  degrees of freedom
##   (2 observations deleted due to missingness)
## AIC: 1139.4
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastSeniorYes ~ factor(PhD01), family = binomial(link = "logit"), 
##     data = polldatacareerstageanalysis)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.8519   0.6300   0.6527   0.6608   0.8758  
## 
## Coefficients:
##                Estimate Std. Error z value Pr(>|z|)    
## (Intercept)     1.41059    0.14281   9.878  < 2e-16 ***
## factor(PhD01)2  0.02738    0.19784   0.138  0.88993    
## factor(PhD01)3  0.10576    0.23296   0.454  0.64983    
## factor(PhD01)4 -0.65000    0.23349  -2.784  0.00537 ** 
## factor(PhD01)5 -0.21434    0.34724  -0.617  0.53706    
## factor(PhD01)6 -0.57226    0.33154  -1.726  0.08434 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1140.7  on 1098  degrees of freedom
## Residual deviance: 1127.4  on 1093  degrees of freedom
##   (2 observations deleted due to missingness)
## AIC: 1139.4
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastSeniorYes ~ PhD01, family = binomial(link = "logit"), 
##     data = polldatacareerstageanalysis)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.8473   0.6330   0.6713   0.7114   0.8424  
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)    
## (Intercept)  1.63639    0.15318  10.683   <2e-16 ***
## PhD01       -0.13049    0.05096  -2.561   0.0104 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1140.7  on 1098  degrees of freedom
## Residual deviance: 1134.3  on 1097  degrees of freedom
##   (2 observations deleted due to missingness)
## AIC: 1138.3
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastSeniorYes ~ WhereLive, family = binomial(link = "logit"), 
##     data = polldatawhereliveanalysis)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.1114   0.4773   0.7860   0.7860   0.7860  
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              2.1151     0.1901  11.125  < 2e-16 ***
## WhereLiveNorth America  -1.0987     0.2081  -5.279  1.3e-07 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1057.8  on 1002  degrees of freedom
## Residual deviance: 1024.7  on 1001  degrees of freedom
##   (2 observations deleted due to missingness)
## AIC: 1028.7
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastSeniorYes ~ ecoevo, family = binomial(link = "logit"), 
##     data = polldataecoevoanalysis)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.2563   0.4042   0.7282   0.7282   0.7282  
## 
## Coefficients:
##                 Estimate Std. Error z value Pr(>|z|)    
## (Intercept)      1.19188    0.07917  15.055   <2e-16 ***
## ecoevoevolution  1.27197    0.52649   2.416   0.0157 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1005.62  on 943  degrees of freedom
## Residual deviance:  997.45  on 942  degrees of freedom
##   (1 observation deleted due to missingness)
## AIC: 1001.5
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastSeniorYes ~ factor(Dept01), family = binomial(link = "logit"), 
##     data = polldata)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.8691   0.6190   0.6319   0.6319   0.8672  
## 
## Coefficients:
##                 Estimate Std. Error z value Pr(>|z|)    
## (Intercept)      1.55516    0.15117  10.288  < 2e-16 ***
## factor(Dept01)2 -0.04554    0.19523  -0.233  0.81554    
## factor(Dept01)3 -0.64876    0.21417  -3.029  0.00245 ** 
## factor(Dept01)4 -0.77104    0.23986  -3.215  0.00131 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1139.9  on 1091  degrees of freedom
## Residual deviance: 1120.5  on 1088  degrees of freedom
##   (30 observations deleted due to missingness)
## AIC: 1128.5
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastSeniorYes ~ BasicApplied, family = binomial(link = "logit"), 
##     data = polldata)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.7721   0.6829   0.6829   0.6829   0.7377  
## 
## Coefficients:
##                   Estimate Std. Error z value Pr(>|z|)    
## (Intercept)         1.1624     0.1235   9.409   <2e-16 ***
## BasicAppliedBasic   0.1746     0.1528   1.142    0.253    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1164.6  on 1110  degrees of freedom
## Residual deviance: 1163.3  on 1109  degrees of freedom
##   (11 observations deleted due to missingness)
## AIC: 1167.3
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastSeniorYes ~ ordered(Inter01), family = binomial(link = "logit"), 
##     data = polldata)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.8170   0.6528   0.6813   0.6856   0.7760  
## 
## Coefficients:
##                    Estimate Std. Error z value Pr(>|z|)    
## (Intercept)         1.25438    0.09486  13.223   <2e-16 ***
## ordered(Inter01).L -0.32772    0.26780  -1.224    0.221    
## ordered(Inter01).Q -0.03991    0.23564  -0.169    0.866    
## ordered(Inter01).C  0.03625    0.18232   0.199    0.842    
## ordered(Inter01)^4  0.07303    0.13962   0.523    0.601    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1167.7  on 1111  degrees of freedom
## Residual deviance: 1164.8  on 1107  degrees of freedom
##   (10 observations deleted due to missingness)
## AIC: 1174.8
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = LastSeniorYes ~ factor(PhD01) + WhereLive + ecoevo + 
##     factor(Dept01) + BasicApplied + factor(Inter01), family = binomial(link = "logit"), 
##     data = polldata)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -2.6847   0.3759   0.5753   0.7159   1.4114  
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)             2.53419    1.16597   2.173   0.0297 *
## factor(PhD01)2         -0.04062    0.21102  -0.193   0.8473  
## factor(PhD01)3          0.07253    0.25733   0.282   0.7780  
## factor(PhD01)4         -0.64578    0.26045  -2.480   0.0132 *
## factor(PhD01)5         -0.29355    0.36578  -0.803   0.4222  
## factor(PhD01)6         -0.74529    0.36574  -2.038   0.0416 *
## factor(PhD01)7         -1.15988    0.52485  -2.210   0.0271 *
## WhereLiveAsia          -0.08813    1.53803  -0.057   0.9543  
## WhereLiveAustralia     -0.52962    1.17525  -0.451   0.6522  
## WhereLiveEurope        -0.06630    1.13870  -0.058   0.9536  
## WhereLiveNorth America -1.14142    1.12360  -1.016   0.3097  
## WhereLiveSouth America -0.59004    1.21063  -0.487   0.6260  
## ecoevoevolution         1.25612    0.53976   2.327   0.0200 *
## ecoevoother             0.42611    0.23144   1.841   0.0656 .
## factor(Dept01)2         0.01599    0.20737   0.077   0.9385  
## factor(Dept01)3        -0.50289    0.23717  -2.120   0.0340 *
## factor(Dept01)4        -0.68132    0.26610  -2.560   0.0105 *
## BasicAppliedBasic      -0.08662    0.18043  -0.480   0.6312  
## factor(Inter01)2       -0.07559    0.31413  -0.241   0.8098  
## factor(Inter01)3       -0.02043    0.30706  -0.067   0.9469  
## factor(Inter01)4       -0.08176    0.32469  -0.252   0.8012  
## factor(Inter01)5       -0.14310    0.44539  -0.321   0.7480  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1119.7  on 1076  degrees of freedom
## Residual deviance: 1043.6  on 1055  degrees of freedom
##   (45 observations deleted due to missingness)
## AIC: 1087.6
## 
## Number of Fisher Scoring iterations: 5

Does whether people pay attention to a CV statement vary based on age, country, research area, department and/or their views on last authorship?

## [1] "Biology other than EEB"                                        
## [2] "Ecology (primarily computational-based)"                       
## [3] "Ecology (primarily field-based)"                               
## [4] "Ecology (primarily wet-lab based, including molecular ecology)"
## [5] "Evolutionary biology (primarily molecular)"                    
## [6] "Evolutionary biology (primarily organismal)"                   
## [7] "Outside biology"

Do views on corresponding authorship vary based on age, country, research area, and/or department?

## Warning in bind_rows_(x, .id): binding character and factor vector,
## coercing into character vector
## Warning in bind_rows_(x, .id): Unequal factor levels: coercing to character
## Warning in bind_rows_(x, .id): binding character and factor vector,
## coercing into character vector

## Warning in bind_rows_(x, .id): binding character and factor vector,
## coercing into character vector

## 
## Call:
## glm(formula = FullResYes ~ ordered(PhD01), family = binomial(link = "logit"), 
##     data = polldatacareerstageanalysis2)
## 
## Deviance Residuals: 
##    Min      1Q  Median      3Q     Max  
## -1.341  -1.235   1.022   1.121   1.226  
## 
## Coefficients:
##                  Estimate Std. Error z value Pr(>|z|)
## (Intercept)       0.09141    0.07841   1.166    0.244
## ordered(PhD01).L -0.26509    0.20717  -1.280    0.201
## ordered(PhD01).Q -0.13931    0.19240  -0.724    0.469
## ordered(PhD01).C  0.13479    0.19783   0.681    0.496
## ordered(PhD01)^4 -0.02353    0.19174  -0.123    0.902
## ordered(PhD01)^5 -0.25566    0.16916  -1.511    0.131
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1514.2  on 1096  degrees of freedom
## Residual deviance: 1508.4  on 1091  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 1520.4
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = FullResYes ~ factor(PhD01), family = binomial(link = "logit"), 
##     data = polldatacareerstageanalysis2)
## 
## Deviance Residuals: 
##    Min      1Q  Median      3Q     Max  
## -1.341  -1.235   1.022   1.121   1.226  
## 
## Coefficients:
##                Estimate Std. Error z value Pr(>|z|)
## (Intercept)     0.13525    0.11367   1.190    0.234
## factor(PhD01)2  0.06954    0.15718   0.442    0.658
## factor(PhD01)3  0.24098    0.18369   1.312    0.190
## factor(PhD01)4 -0.22487    0.20696  -1.087    0.277
## factor(PhD01)5 -0.10016    0.28830  -0.347    0.728
## factor(PhD01)6 -0.24858    0.29772  -0.835    0.404
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1514.2  on 1096  degrees of freedom
## Residual deviance: 1508.4  on 1091  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 1520.4
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = FullResYes ~ PhD01, family = binomial(link = "logit"), 
##     data = polldatacareerstageanalysis2)
## 
## Deviance Residuals: 
##    Min      1Q  Median      3Q     Max  
## -1.273  -1.253   1.085   1.103   1.177  
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)  
## (Intercept)  0.26472    0.12397   2.135   0.0327 *
## PhD01       -0.04387    0.04331  -1.013   0.3111  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1514.2  on 1096  degrees of freedom
## Residual deviance: 1513.1  on 1095  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 1517.1
## 
## Number of Fisher Scoring iterations: 3
## 
## Call:
## glm(formula = FullResYes ~ WhereLive, family = binomial(link = "logit"), 
##     data = polldatawhereliveanalysis2)
## 
## Deviance Residuals: 
##    Min      1Q  Median      3Q     Max  
## -1.311  -1.210   1.050   1.145   1.145  
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)   
## (Intercept)              0.3080     0.1192   2.583  0.00981 **
## WhereLiveNorth America  -0.2322     0.1408  -1.649  0.09924 . 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1382.6  on 1000  degrees of freedom
## Residual deviance: 1379.9  on  999  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 1383.9
## 
## Number of Fisher Scoring iterations: 3
## 
## Call:
## glm(formula = FullResYes ~ ecoevo, family = binomial(link = "logit"), 
##     data = polldataecoevoanalysis2)
## 
## Deviance Residuals: 
##    Min      1Q  Median      3Q     Max  
## -1.266  -1.266   1.091   1.091   1.228  
## 
## Coefficients:
##                 Estimate Std. Error z value Pr(>|z|)   
## (Intercept)      0.20701    0.06732   3.075  0.00211 **
## ecoevoevolution -0.32480    0.28851  -1.126  0.26026   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1298.9  on 942  degrees of freedom
## Residual deviance: 1297.6  on 941  degrees of freedom
##   (2 observations deleted due to missingness)
## AIC: 1301.6
## 
## Number of Fisher Scoring iterations: 3
## 
## Call:
## glm(formula = FullResYes ~ factor(Dept01), family = binomial(link = "logit"), 
##     data = polldatadepttypeanalysis2)
## 
## Deviance Residuals: 
##    Min      1Q  Median      3Q     Max  
## -1.331  -1.174   1.031   1.090   1.181  
## 
## Coefficients:
##                 Estimate Std. Error z value Pr(>|z|)   
## (Intercept)       0.3535     0.1167   3.029  0.00245 **
## factor(Dept01)2  -0.3625     0.1506  -2.408  0.01605 * 
## factor(Dept01)3  -0.1452     0.1808  -0.803  0.42202   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1321.1  on 956  degrees of freedom
## Residual deviance: 1315.1  on 954  degrees of freedom
##   (3 observations deleted due to missingness)
## AIC: 1321.1
## 
## Number of Fisher Scoring iterations: 4
## 
## Call:
## glm(formula = FullResYes ~ factor(PhD01) + WhereLive + ecoevo + 
##     factor(Dept01), family = binomial(link = "logit"), data = polldata)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.6845  -1.2146   0.9402   1.0989   1.4406  
## 
## Coefficients:
##                         Estimate Std. Error z value Pr(>|z|)  
## (Intercept)             1.346343   0.837631   1.607   0.1080  
## factor(PhD01)2          0.089420   0.162309   0.551   0.5817  
## factor(PhD01)3          0.241222   0.191887   1.257   0.2087  
## factor(PhD01)4         -0.215497   0.215907  -0.998   0.3182  
## factor(PhD01)5         -0.087409   0.294602  -0.297   0.7667  
## factor(PhD01)6         -0.267517   0.308385  -0.867   0.3857  
## factor(PhD01)7         -0.008312   0.492158  -0.017   0.9865  
## WhereLiveAsia          -0.875818   1.014683  -0.863   0.3881  
## WhereLiveAustralia     -0.976865   0.865204  -1.129   0.2589  
## WhereLiveEurope        -0.848316   0.834541  -1.017   0.3094  
## WhereLiveNorth America -1.043690   0.828131  -1.260   0.2076  
## WhereLiveSouth America -0.833515   0.906331  -0.920   0.3578  
## ecoevoevolution        -0.340404   0.299860  -1.135   0.2563  
## ecoevoother            -0.207584   0.172915  -1.200   0.2299  
## factor(Dept01)2        -0.347022   0.153571  -2.260   0.0238 *
## factor(Dept01)3        -0.150567   0.186054  -0.809   0.4184  
## factor(Dept01)4        -0.204547   0.216299  -0.946   0.3443  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1501.8  on 1087  degrees of freedom
## Residual deviance: 1483.1  on 1071  degrees of freedom
##   (34 observations deleted due to missingness)
## AIC: 1517.1
## 
## Number of Fisher Scoring iterations: 4